Bioprocessing strategies for cost-effective simultaneous removal of chromium and malachite green by marine alga Enteromorpha intestinalis

A large number of industries use heavy metal cations to fix dyes in fabrication processes. Malachite green (MG) is used in many factories and in aquaculture production to treat parasites, and it has genotoxic and carcinogenic effects. Chromium is used to fix the dyes and it is a global toxic heavy metal. Face centered central composite design (FCCCD) has been used to determine the most significant factors for enhanced simultaneous removal of MG and chromium ions from aqueous solutions using marine green alga Enteromorpha intestinalis biomass collected from Jeddah beach. The dry biomass of E. intestinalis samples were also examined using SEM and FTIR before and after MG and chromium biosoptions. The predicted results indicated that 4.3 g/L E. intestinalis biomass was simultaneously removed 99.63% of MG and 93.38% of chromium from aqueous solution using a MG concentration of 7.97 mg/L, the chromium concentration of 192.45 mg/L, pH 9.92, the contact time was 38.5 min with an agitation of 200 rpm. FTIR and SEM proved the change in characteristics of algal biomass after treatments. The dry biomass of E. intestinalis has the capacity to remove MG and chromium from aquatic effluents in a feasible and efficient manner.


Results and discussion
The biosorption processes are complicated systems and their performance is greatly affected by various physicochemical process parameters such as pH, temperature, etc. In this study, the effects of five factors, namely biomass of E. intestinalis as a biosorbent, the concentration of chromium ions, the concentration of MG dye, initial pH level and the contact time on the removal efficiency of chromium ions and MG dye (as responses) were evaluated.

Statistical optimization of chromium and MG removal by E. intestinalis biomass.
A total number of fifty experimental trials of FCCCD (Table 1) were used to evaluate the impacts of five process variables and to determine their optimal levels for simultaneous removal of chromium and MG from aqueous solutions when the malachite green concentration was 6 mg/L, chromium concentration was 120 mg/L, algal biomass was 3 g/L, initial pH level was 10 and the incubation time was 40 min. While the minimum chromium (47.14%) and malachite green (7.79%) removal obtained in the run no. 2 when the malachite green concentration was 2 mg/L, chromium concentration was 120 mg/L, algal biomass was 3 g/L, initial pH level was 7 and the incubation time was 40 min.
Multiple regression analysis and ANOVA. The results of FCCCD for removal of malachite green by E.
intestinalis biomass were analyzed by multiple regression statistical analysis and ANOVA (analysis of variance) calculations which are tabulated in Table 2. Statistical regression analysis parameters such as determination coefficient (R 2 ) value, predicted R 2 value, adj R 2 value, F-value and lack of fit have been determined and evaluated for the model reliability.
A regression model with a value of R 2 exceeding 0.9 was considered strongly correlated 34 . The current R 2 value of the model used for malachite green removal by E. intestinalis (R 2 = 0.9888) reflects that 98.88% of variance in malachite green removal were assigned to the used factors and the model cannot explain just 1.22 per cent of the total variance. In addition, the Adj R 2 value of the malachite green removal % (Adj R 2 = 0.9810) was high also to verify the great model significance ( Table 2). The value of predicted R 2 of 0.9642 agreed with the value of the Adj R 2 . This indicates a strong correlation between the experimental and predicted values of the malachite green removal percentages. A relatively small value of the coefficient of variation % (C.V. = 6.77%) reflects high precision and accuracy of the experiments values 35 (Table 2). Here, the ANOVA for the malachite green removal % indicate that the model terms are highly significant which is confirmed by the F (Fishers' variance ratio) value (F-value = 127.64) and a very small P-value [˂ 0.0001] ( Table 2). P-value less than 0.05 indicate that the terms of the model are significant 36 . The lack of fit for malachite green removal % is not significant (F-value = 3.02; P-value = 0.0688) ( Table 2).
Data were interpreted by means of the signs of the coefficients (negative or positive impact on the response) and P-value (P < 0.05) for understanding the interactions between test variables. Two-factor interactions can appear as an oppositional (negative) or complementary (positive) effect. The significance value of coefficients can indicate that the linear coefficients of X 1 , X 2 , X 4 and X 5 are highly significant together with the interaction effects between X 1 X 2 , X 1 X 3 , X 1 X 4 , X 1 X 5 , X 2 X 4 , X 3 X 5 , X 2 X 5 , X 2 1 , X 2 3 , X 2 4 and X 5 2 . In addition, the P-value of coefficients (P-value < 0.05) can indicate that the interactions between X 1 and X 2 ; X 1 X 5 ; X 3 X 5 had a very significant impact on malachite green decolourization by E. intestinalis. The linear coefficients of X 3 , interactions between X 2 X 3 , X 3 X 4 and X 4 X 5 and X 2 quadratic effect are nonsignificant model terms that do not make a significant contribution to the malachite green removal.
The fit summary results seen in Supplementary Table S1 indicate that the quadratic polynomial model is the highest significant model and sufficient to fit the FCCCD of malachite green removal by E. intestinalis where the terms are significant (P-value < 0.0001) with non-significant lack of fit (P-value = 0.0688; F-value = 3.02). The quadratic model summary data indicate the lower Std. Dev. value (4.12) and higher values of the adjusted and predicted R 2 (0.9810 and 0.9642; respectively).
The polynomial regression equation of second order for malachite green removal by E. intestinalis (Y) can be written according to the coefficients that were fitted as the following: where Y is the predicted value of malachite green removal % by E. intestinalis biomass. X 1 -X 5 are coded values for the concentration of malachite green, chromium concentration, E. intestinalis biomass concentration, initial pH level and contact time.
Similarly, the results of FCCCD for chromium ions removal % by E. intestinalis biomass were analyzed by multiple regression statistical analysis and ANOVA (analysis of variance) calculations which are tabulated in www.nature.com/scientificreports/  (Table 3). Here, the ANOVA of the quadratic regression model for the chromium ions removal % verify that the model terms are highly significant which is confirmed by the F (Fishers' variance ratio) value (F-value = 199.94) and a very small P-value [˂ 0.0001] ( Table 3). The lack of fit for chromium ions removal % is not significant (F-value = 1.91; P-value = 0.1927) ( Table 3). The significance value of coefficients can indicate that all the linear and quadratic coefficients are significant. The coefficients P-values also indicate that between the five factors studied, two-factor interactions between X 1 , X 2 (MG conc. and chromium conc.), X 1 X 3 (MG conc. and algal biomass conc.), X 1 X 5 (MG conc. and incubation time), X 2 X 4 (chromium conc. and initial pH ), X 2 X 5 (chromium conc. and incubation time), X 3 X 4 (algal biomass and initial pH) and X 3 X 5 (algal biomass and incubation time ) had a very significant effects on chromium removal by E. intestinalis. On the other hand, the interactions between X 1 X 4 ; X 2 X 3 ; X 4 X 5 are no significant model terms that do not make a significant contribution to the removal of chromium ions.
The fit summary results seen in Supplementary Table S2 show that the quadratic polynomial model is the highest significant and sufficient to fit the FCCCD of chromium ions removal by E. intestinalis where the terms are significant (P-value < 0.0001) and lack of fit is not significant (P-value = 0.1927; F-value = 1.91).
The polynomial regression equation of second order for chromium ions removal by E. intestinalis (Y) can be written according to the coefficients that were fitted as the following: where Y is the predicted value of chromium ions removal % by E. intestinalis biomass. X 1 -X 5 are coded values for the concentration of malachite green, chromium concentration, E. intestinalis biomass concentration, initial pH level and contact time.
(2) www.nature.com/scientificreports/ Three dimensional (3D) plots for malachite green removal. The 3D graphs are tools to understand the interactions between the process factors and to predict the optimal conditions for the highest percentage of malachite green removal. 3D graphs for the five variables combined in pairs "X 1 X 2 , X 1 X 3, X 1 X 4, X 1 X 5, X 2 X 3, X 2 X 4, X 2 X 5, X 3 X 4, X 3 X 5, and X 4 X 5 " were constructed by plotting the percentages of malachite green removal on Z-axis versus two independent process factors while maintaining the other independent process factors at their center levels. The 3D graph (Fig. 1A), shows the impact of malachite green concentration (X 1 ) and chromium concentration (X 2 ) on the percentage of malachite green removal, whereas E. intestinalis biomass concentration (X 3 ), initial pH (X 4 ) and incubation time (X 5 ) were maintained their center levels. Figure 1A indicates that the highest percentage of malachite green removal is obviously located close to the central level of malachite green concentration. Furthermore, the lower and higher concentrations of malachite green (X 1 ) resulted in lower malachite green removal percentages. By analyzing Fig. 1A and solving the Eq. (1), the maximum predicted value for malachite green removal of 97.70% could be attained at the optimal predicted levels of malachite green and chromium concentrations of 10 and 200 mg/L; respectively by using E. intestinalis biomass concentration of 3 g, initial pH 7 and 40 min incubation time.
The 3D graph (Fig. 1B), showing the effects of malachite green concentrations (X 1 ) and E. intestinalis biomass concentrations (X 3 ) on the percentage of malachite green decolourization, at center levels of chromium concentrations (X 2 ), initial pH (X 4 ) and incubation time (X 5 ). Figure 1B indicates that the highest percentage of malachite green removal was attained by using 3 g/L E. intestinalis biomass concentration, after which the decolourization of malachite green decreased. The lower and higher concentrations of malachite green (X 1 ) resulted in low percentage of malachite green decolourization and the highest percentage of malachite green removal obviously located at center levels of malachite green. By analyzing Fig. 1B and solving the Eq. (1), the maximum predicted malachite green removal of 97.07% could be attained at the optimal predicted levels of malachite green and E. intestinalis biomass concentrations of 6 mg/L and 3 g/L; respectively by using chromium concentrations of 120 mg/L, initial pH 7 and 40 min contact time.
The 3D graph (Fig. 1C), showing the effects of two factors, malachite green concentrations (X 1 ) and initial pH level (X 4 ), on malachite green removal percentage, while the other factors (chromium concentrations, E. intestinalis biomass concentration and contact time) were kept at their center levels. The percentage of malachite green removal increased gradually with increasing levels of malachite green concentrations to the central level, www.nature.com/scientificreports/ after which the malachite green removal decreased. On the other hand, 3D graph (Fig. 1C), indicates that the high levels of initial pH increased malachite green decolourization. By analyzing Fig. 1C and solving the Eq. (1), the maximum predicted malachite green removal of 97.7% could be attained at the optimal predicted levels of 6 mg/L malachite green (X 1 ) and pH 8 by using 120 mg/L chromium concentration, 3 g/L E. intestinalis biomass concentration and 40 min contact time. www.nature.com/scientificreports/ The 3D graph (Fig. 1D), showing the effects of malachite green concentrations (X 1 ) and contact time (X 5 ) on the malachite green decolourization efficiency, when the chromium concentrations (X 2 ), E. intestinalis biomass concentration (X 3 ) and initial pH (X 4 ) were kept at their center levels. By analyzing Fig. 1D and solving the Eq. (1), the maximum predicted malachite green removal of 97.7 percent could be attained at the optimal predicted levels of 6.5 mg/L malachite green (X 1 ) and contact time (X 5 ) of 45 min by using 120 mg/L chromium concentration, 3 g/L E. intestinalis biomass concentration and pH 7.
The 3D plots (Fig. 1E-G) represent the effects of chromium concentrations (X 2 ) and algal biomass (X 3 ) (Fig. 1E); chromium concentrations (X 2 ) and pH (X 4 ) (Fig. 1F); chromium concentrations (X 2 ) and contact time (X 5 ) (Fig. 1G) on the malachite green decolourization efficiency, when the other independent variables were kept at their center levels. Figure 1E-G shows that the lower and higher levels of chromium concentrations, algal biomass and contact time led to a low percentage of malachite green removal while, the higher level in pH support increase in the malachite green decolourization.
The three-dimensional response surface curves in Fig. 1H,I indicates that the higher and lower levels of alga biomass increase the malachite green decolourization but the higher level of pH causes increase in malachite green decolourization. Figure 1J showed lower and higher levels of contact time decrease malachite green removal percentage and higher value of malachite green decolourization was obtained beyond high pH value.
The adequacy of the model. The normal probability plot is the graph that signifying the normal distribution of the residuals to validate the model suitability 37 . The residuals are the differences between the responses' experimental values and their predicted theoretical values. Low residual values indicate very accurate model prediction 38 . Figure 2A shows the studentized residuals plotted versus the normal probability for malachite green removal efficiency by E. intestinalis biomass. The residuals are normally distributed; they are located along www.nature.com/scientificreports/ the straight diagonal line of malachite green decolourization %. Therefore, the normal distribution of the residuals reveals the model's validity 39 . Figure 2B shows the actual versus predicted percentages for malachite green removal percentages from aqueous solution. Figure 2B displays all the points along the diagonal line, indicating that the model's predicted percentages coincide with the actual percentages, confirming that the model is accurate. Figure 2C shows the studentized residual versus predicted values for malachite green removal percentages. Figure 2C in this study indicated that the residuals randomly distributed about zero line. This meant that the residuals had an almost constant variance over the variable ranges. Figure 2D shows Box-Cox plot of model transformation of malachite green removal percentages. As can be seen in Fig. 2D, the Lambda (λ) optimal value of 1 lies between the two vertical red lines so that no data transformation is required.
Three dimensional (3D) plots for chromium removal. Figure 3 presents the three-dimensional plot for chromium removal percentages as a function of malachite green concentration, chromium concentrations, algal biomasses, initial pH level and incubation time. Figure 3A-D demonstrates that higher and lower levels of malachite green decrease the percentage of chromium removal from aqueous solutions and the maximum chromium removal percent attained at the middle level of malachite green. Figure 3A,B demonstrates that the lower and higher levels of algal biomass increase the chromium removal percentage; Fig. 3C, higher levels of pH and middle levels of malachite green concentrations causes an increase of chromium removal percentage. Figure 3D reveals that the contact time has a low effect the percentage of chromium removal. The 3D plots obtained in Fig. 3E-G presents the effects of independent variable chromium concentrations and algal biomass (Fig. 3E); chromium concentrations and pH (Fig. 3F); chromium concentrations and contact time (Fig. 3G). The 3D plots indicated that chromium removal percentage increased at the central (zero) levels of biomass (Fig. 3E), at central levels of contact time (Fig. 3G), the high and low values of contact time resulted in a chromium removal decrease. Figure 3F shows that the maximum chromium removal % has been attained at the central level of chromium concentrations and at high pH level. Figure 3H,I depicted that the effect of independent variable, algal biomass, contact time and pH, while the other variables were kept at their center levels. The percentage of chromium removal was decreased at low and high levels of algal biomass and contact time and increase with an increase in pH.
The removal of different dyes and metal ions increased with increasing the dye and metal ions solely or simultaneously and reached to highest. Further increasing in dye and metal ions concentrations leads to a slight increase in the removal percentage. This can be due to all active sites on the algal biomass adsorptive for metals ions and MG that free at the beginning resulting in high dye and metals ion adsorptions, so further increasing of MG and heavy metals ions resulting in decreasing of adsorption due to algal biomass free active site are few to binding with excess MG dye or metal ions, So dry algae can adsorbed heavy metals and dyes effectively, but it was most affected and limited by optimization of the required process factors such as temp., pH, algal biomass, as well as, the concentrations of dyes and heavy metals. Husien et al. 40 reported that when the concentrations of pollutant increased the removal of pollutants was decreased due to the binding sites available were decreased on the surface of algae.
Effect of chromium concentrations. The chromium concentrations are the most important factors that impact of chromium removal by algae, so chromium removal was decreased by increasing chromium concentrations due to the Chlorella cells was degraded 40,41 . Al-Homaidan et al. 32 reported that the removal rate of chromium increase for initial concentrations of chromium in range 10 to 20 mg/L but decrease above this level due to binding sites saturation. The numerous number of chromium ions was competing with the binding sites of the algal biomass 42 . Sutkowy and Klosowski 43 applied the alga Pseudopediastrum sp. as biosorpent of Cr (VI), they reported that the biosorption capacity when increasing of initial concentrations of the metal. Kumar et al. 44 reported that increase of initial concentrations of Cr(VI) resulted in an increase in chromium sorption by filamentous algae that may be due to the accessibility of more surface area of the adsorbent. When the concentration of chromium increases, Chlorella vulgaris and Scenedesmus acutus remove the least amount of chromium despite increasing the driving force 45 . Zhang et al. 46 reported that the chromium removal by activated carbon derived from algal bloom residue were decreased from 91.9 to 85.5% with increasing initial Cr(VI) concentration from 50 to 200 mg/L. The chromium removal capacity by brown alga Dictyopteris polypodioides decreased from 96.3 to 37.6%, when chromium concentration increased from 50 to 500 mg/L due to the binding sites saturation 47 . Li et al. 48 investigated that the uptake of Cr(VI) by Polysiphonia urceolata was ranged from 16.1 to 128.2 mg/L and by Chondrus ocellatus was ranged from 17.3 to 105.2 mg/L when chromium concentrations varied from 25 to 250 mg/L, the increase of percentage removal may be due to increase of biosorbent doses. Katircioğlu et al. 49 used Oscillatoria sp. as biosorbent for Cr (VI), and demonstrated that the chromium removal increased when the initial Cr(VI) concentration was increased from 25 to 200 mg/L.

Effect of malachite green concentrations. The percentage of decolonization of malachite green by
Enteromorpha was decreased with increase dye concentrations 31 . The removal percentage of malachite green by algal bloom residues decreased from 50.9 to 33.9% when the initial concentration of MG was increased from 50 to 100 mg/L 50   www.nature.com/scientificreports/ There is a direct relationship between negative charge and pH; an increase in pH causes an increase in negative charge of functional groups until all functional groups are deprotonated 58 . Data collected in Table 4 clear that the maximum removal of MG was at pH ranged from 5 to 10 when using different algae as adsorbent 52,54,57,[59][60][61][62][63][64][65] . Also in agreement with study, the maximum removal of MG by algae Sargassum crassifolium, Gracilaria corticata and Turbinaria conoides was obtained at pH 8. On the other hand, the maximum removal of MG by Ulva lactuca was obtained at pH 7 65 . The maximum removal of the MG by Sargassum swartzii was obtained at pH 10 66 .
According to the summarized data in Table 5, the optimum pH, initial chromium concentrations and also initial adsorbent concentrations vary according to the algae types 40,[43][44][45]48,49,[67][68][69][70] . With an increase in pH, the number of negatively charged binding sites increases, which would attract more cations (positive charge) of heavy metals or basic dye (MG) 71 . So in this study the optimum pH was 9.92 for simultaneous removal of MG and chromium ions.
Effect of the biosorbent dosage. In this study, the biosorbent dosage (Enteromorpha biomass concentration) affects the removal efficiency of both MG and chromium. The highest removal efficiency of both MG and chromium was obtained using 4.3 g/L of Enteromorpha biomass concentration. A highest removal of chromium was 66.6% when using 1.0 g of the dried alga, Cladophora glomerata, /100 mL aqueous solutions contains 20 mg/L chromium 32 . The highest chromium removal percentage (99.75%) by dry alga, Chlorella vulgaris, was obtained using 60 mg/50 mL solutions (1.2 g/L) 70 . Gandhi et al. 72 demonstrated that the highest percentage uptake of chromium was obtained with 8.0 g algae as biosorbent. The highest chromium removal (83.55%) was obtained with 0.6 g/L Sargassum sp. after 120 min of contact time 73 . Whereas, highest chromium removal was obtained with 60 mg/L Sargassum sp. after 40 min of contact time 74 .  www.nature.com/scientificreports/ Effect of contact (incubation) time. The maximum dye removal (91.92%) was obtained by using 1.25 g/L U. lactuca as biosorbent after 110 min of contact time 75 . The maximum removal of MG by Scenedesmus sp. MCC26 was obtained after 60 min of contact time 76 . Al-Homaidan et al. 32 reported that the removal of chromium by green algae (Microspora amoena, Enteromorpha intestinalis and Cladophora glomerata) remain constant after one hour which indicated saturations. Gurbuz 77 noticed that the removal of Cr(VI) ions when using green alga Scenedesmus as biosorbent was quick during the first 30 min (65.62 ± 2.4%), then increase to 92.7 ± 4.12% after 1 h. Sala et al. 78 reported that the maximum removal of chromium ions (60%) by dried marine alga Sargassum sp. was obtained in ten min.
Optimization using the desirability function. Design Expert software was used for optimization to identify the best working conditions for the highest simultaneous malachite green and chromium ions removal. The program's desirability function has been set from zero to one for each factor. The maximization of this desirability function is the ultimate objective of this program. Due to the curvature format of the response surfaces, more than one maximum point is expected, and their combinations into the desirability function. This software begins in the design space from many points, until the search completes by finding the best maximum for the responses 79,80 . Figure 4 shows the desirability values of the numerical optimization to find the optimum points which maximizes the removal % of both malachite green removal and chromium ions. Figure 4 shows that the maximum predicted malachite green removal and chromium removal could be 99. 63  FTIR analysis. The FTIR spectrums of E. intestinalis biomass samples were analyzed before and after biosorption of malachite green and chromium ( Table 6, Fig. 5) to notice any differences because of the interaction of dye and metal ions with binding sites (functional groups) that occurs on the biomass cell surface. "The macro green alga cell walls, consist of the major content of polysaccharides, and have many functional groups which carrying negative charges that can interact with cationic dye and bind heavy metal ions, and these functional groups include carboxylate, hydroxyl, amino and phosphate groups 81 . The spectra of adsorbents before and after treatments were measured within the range of 400-4,000 cm −1 wave number" 82 .The spectrum of FTIR analysis   87 . The peaks at 1,034 cm -1 correspond to the C-N stretching mode 88 . Peak at 851 referred to C(1)-H(α) bending 89 . Peaks ranged from 900 to 675 (s) assigned to C-H "oop" aromatics 90 . After the malachite green and heavy metals absorption the wavenumber of the peaks are shifted to higher or less wavenumber. The -OH absorption peak at 3,448 cm −1 is shifted to 3,508, 3,483 and 3,450 cm −1 . There are two small peaks at 2,293 cm −1 and 2,266 cm −1 observed in the FT-IR spectroscopy curve after absorption of malachite green and chromium; these peaks may due to alkanes 91 . Figure 5 demonstrated that peaks 1653, 1,459, 1,034, 851 and 798 cm −1 are shifted to 1,650, 1,426, 1,261, 1,031, 849 and 797 cm −1 . These shifted absorption peaks could be attributable to chemical bonding among binding sites on algal biomass and the malachite green dyes, or chromium 92 . The small difference between the wave number of peaks after and before treatments with simultaneously malachite green and chromium it is presumed that the dye and heavy metals incorporated within the adsorbent through interaction with the active functional groups 31 .
Scanning electron microscopy (SEM). Figure 6A,B shows that SEM micrograph of E. intestinalis biomass after and before malachite green and chromium adsorption. The results investigated that the control alga is relatively smooth surface and a little amount of impurities was present, whereas the treated alga had a rough surface and present a large amount of impurities may be due to MG and chromium absorbed on the alga surface. The cell wall of Sargassum swartzii after biosorption of MG appeared shrinkage in comparison to alga before absorption MG 66,93 . The rough surface with micropores of Chlorella vulgaris particles was showed under SEM after absorption of chromium 70 .

Material and methods
Collection and preparation of biosorbent. E. intestinalis was collected from Jeddah, Saudi Arabia beach on April 2019, and was identified according to Taylor 94 . The E. intestinalis biomass were washed thoroughly under running tap water, and then distilled water to completely remove salts and sand. The cleaned marine green alga biomass was dried in oven at 60 °C, until the moisture was completely removed (up to constant weight). Furthermore, the dried alga biomass was milled and the grounded powder was sieved using the standard laboratory test sieve. The ground alga biomass with particle size range of 1-1.2 mm was used as biosorbent for biosorption experiments for simultaneously malachite green and chromium removal in the present study 8 .
Preparation of malachite green and heavy metal solutions. The required solutions used for the biosorption experiments were prepared. The initial concentrations of Cr(VI) ions (40, 120, 200 mg/L) and malachite green (2, 6, 10 mg/L) were prepared by dissolving known quantity of potassium dichromate or malachite green in 1 L distilled water 95,96 . The initial pH level of each solution was adjusted using 0.1 N HCL and 0.1 N NaOH to the desired level. www.nature.com/scientificreports/

Statistical optimization of chromium and Malachite green biosorption by face centered central composite design (FCCCD).
The biosorption experiments were conducted in batch condition at room temperature (28 ± 2°C). The biosorption experiments were carried out in a series of 250 mL Erlenmeyer flask using FCCCD to evaluate the impacts of five variables and to determine their optimal levels on the chromium and MG biosorption. Fifty experimental trials which are shown in Table 1 were conducted with 8 runs at the midpoint for five process variables and each variable varies in three levels: − 1 (low level), 0 (standard, middle or zero level) and + 1 (high level). The chosen independent variables were initial concentration of MG (X 1 ; 4, 6, 10 mg/L), initial concentration of Cr(VI) (X 2 ; 40, 120, 200 mg/L), biosorbent concentration (X 3 ; 1, 3, 5 g/L), initial pH level (X 4 ; 4, 7, 10) and contact time (X 5 ; 20, 40, 60 min) at a constant agitation speed (200 rpm). The relationships between the five independent process variables and the responses (% Cr(VI) and MG biosorption) were determined using the second-degree polynomial equation as follows: www.nature.com/scientificreports/ In which Y is the predicted Cr(VI) or MG biosorption perecntage, the linear coefficient (β i ), quadratic coefficients (β ii ), the regression coefficients (β 0 ), the interaction coefficients (β ij ) and the coded values of the independent variables (X i ).
Statistical analysis. Design Expert version 7 for Windows software was used for the experimental designs and statistical analysis. The statistical software package, STATISTICA software (Version 8.0, StatSoft Inc., Tulsa, USA) was used to plot the three-dimensional surface plots.
Analytical methods. Ten milliliters of the binary solution for each trial of FCCCD was centrifuged, and the supernatants were analyzed using the spectrophotometer by measuring the absorbance changes at a wavelength of λ max 616 nm to determine the final (residual) concentrations (Cf) of malachite green dye. The efficiency of E. intestinalis biomass for malachite green removal from aqueous solutions was determined in percentage using the following equation: where: C i , C f are the initial and final malachite green concentrations (mg/L); respectively. Another 10 mL of the binary solution for each trial were analyzed to determine the residual concentration of Cr(VI) ions using Atomic absorptions (Buck scientific 2 Accusystem series Atomic Absorption (USA) by an air acetylene system) in the Biotechnology Unit, Mansoura university Egypt 97 . The efficiency of E. intestinalis biomass for chromium ions elimination from aqueous solutions was determined in percentage using the following equation: www.nature.com/scientificreports/ where: C i , C f are the initial and final chromium ions concentrations (mg/L); respectively. All determinations of both chromium ions and malachite green in the binary solution were estimated in triplicates.
Fourier transform infrared (FTIR) spectroscopy. The FTIR spectroscopy is a significant tool used to identify the distinctive functional groups that may be responsible for the biosorption process of both malachite green and chromium ions by the surface of E. intestinalis biomass. The dry biomass of E. intestinalis samples were analyzed using FTIR spectroscopy before and after malachite green and chromium ions removal. The samples of dry biomass were mixed with pellets of potassium bromide and the FTIR spectra were then determined within the range of 400-4,000 cm −1 using "Thermo Fisher Nicolete IS10, USA spectrophotometer".
Scanning electron microscopy (SEM). The samples of E. intestinalis dry biomass were investigated after and before chromium and malachite green removal using SEM to examine the cell surface morphology of E. intestinalis biomass before and after the biosorption process of both chromium and malachite green. The gold-coated dry biomass samples were investigated at various magnifications using accelerated beam voltage of 30 keV.